Python common commands in deep learning

1. Print Dafa

Test = Hello worldprint ("Test:" + test)

2. The difference between math and numpy: math is only for a single element, NumPy broadcasting.

Import Mathimport numpy as npx = [1, 2, 3]s = 1/(1+math.exp (-X)  #这条语句会报错s = 1/(1+np.exp (x)) #这条语句没问题.

3. Defining functions

def sigmoid_derivative (x):    s = 1/(1+np.exp (-X)    ds = s* (1-s)    return DSX = Np.array ([1, 2, 3]) print ("Sigmoid_d Erivative (x) = "+ str (sigmoid_derivative (x))

4. Shape and reshape

image = Np.array ([[[0.67826139, 0.29380381], [0.90714982, 0.52835647], [0.4215251, 0.45017551]], [[0.92814219, 0.96677647], [0.85304703, 0.52351845], [0.199813 97, 0.27417313]], [[0.60659855, 0.00533165], [0.10820313, 0.49978937], [0.34144279, 0.94630077] ], [[0.85304703, 0.52835647], [0.10820313, 0.45017551], [0.34144279, 0.9071498 2]]) print ("image[3][2][1]:" + str (image[3][2][1)) # Image[2][3][1]: 0.90714982print ("image.shape =" + str (image.shap  e) # Image.shape = (4, 3, 2) vector = Image.reshape (image.shape[0]*image.shape[1]*image.shape[2], 1) print ("Vector.shape = "+ str (vector.shape)) # Vector.shape = (1) 

5. Normalize:x_norm = Np.linalg.norm (x, ord = 2, Axis = 1, keepdims = True), where ord=2 is the default value can not write, Axis=1 is the horizontal amount normalization, for one dimensional vector, axis can only For 0,keepdims=true is to keep the shape of the array to prevent the appearance (2,) of this shape, just in case as far as possible write Keepdims=true.

x = Np.array ([[0,3,4],[2,6,4]]) X_norm = Np.linalg.norm (x, ord=2, Axis =1, keepdims=true) x_new = X/x_normprint ("x:" + str (x)) Print ("X_norm:" +str (X_norm)) print ("X_new:" +str (x_new)) output: x: [[[0] 3 4] [2 6 4]]x_norm: [[5        ] [7.48331477]]x_new: [[0.          0.6         0.8       ] [0.26726124  0.80178373  0.53452248]]

6. Summation: X_sum = np.sum (x, Axis = 1, keepdims = True), where Axis=1 is the sum of the horizontal quantities.

x = Np.array ([[0,3,4],[2,6,4]]) X_sum = np.sum (x, Axis = 1, keepdims = True) print ("X_sum:" +str (x_sum)) output: x_sum: [[7] [12 ]]

7. Different multiplication:

Np.dot (x1, x2) is normal matrix multiplication for matrices, and for one-dimensional vectors it is the corresponding element multiplied and summed;

Np.multiply (x1, x2) is a one-dimensional vector that is multiplied by the corresponding elements of a single dimension.

Here time is the method of timing.

Import TIMEX1 = [9, 2, 5, 0, 0, 7, 5, 0, 0, 0, 9, 2, 5, 0, 0]x2 = [9, 2, 2, 9, 0, 9, 2, 5, 0, 0, 9, 2, 5, 0, 0]### vector point multiplication, corresponding Element multiplication and Summation # # #tic = time.process_time () dot = Np.dot (x1,x2) TOC = Time.process_time () print ("dot =" + str (dot) + "\ n-----Com Putation time = "+ str (1000* (toc-tic)) +" MS ") # # # X1 and x2 transpose do matrix multiplication, n*1 matrix multiplied by 1*n Matrix # # #tic = time.process_time () outer = NP.O Uter (x1,x2) TOC = Time.process_time () print ("outer =" + str (outer) + "\ n-----Computation time =" + str (1000* (toc-tic) + "MS") # # # corresponding element multiplied to get 1*n vector # # #tic = Time.process_time () Mul = np.multiply (x1,x2) TOC = Time.process_time () print ("Elementw Ise multiplication = "+ str (mul) +" \ n-----Computation time = "+ str (1000* (toc-tic)) +" MS ") # # # normal Matrix multiplication # # #W = Np.ra Ndom.rand (3,len (x1)) # Random 3*len (x1) NumPy arraytic = time.process_time () dot = Np.dot (w,x1) TOC = Time.process_time () pr int ("GDOT =" + str (dot) + "\ n-----Computation time =" + str (1000* (toc-tic)) + "MS") output: dot = 278-----computation T IME = 0.0msouter = [[81 18] 18 81 0 81 18 45 0 0 81 18 45 0 0 "[18 4 4 18 0 18 4 10 0 0 18 4 10 0 0] [45 10 10 45 0 45 10 25 0 0 45  10 25 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [63 14 14 63 0  63 14 35 0 0 63 14 35 0 0] [45 10 10 45 0 45 10 25 0 0 45 10 25 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [81 18 18 81 0 81 18 45 0 0 81 18 45 0 0] [18 4 4 18 0 18 4 10 0 0 18 4 10 0 0] [45 10 10 45 0 45 10 25 0 0 45 10 25 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]] -----computation time = 0.0mselementwise multiplication = [Bayi 4 0 0 0 0 0 Bayi 4, 0 0]-----computatio N time = 0.0msgdot = [14.98632469 18.30746169 17.30396991]-----computation time = 0.0ms

8. Broadcasting:loss = Np.sum ((yhat-y) **2, keepdims = True), the operation of this * *, also calculates the square for each element.

　　

　　

　　

　　

　　

Python common commands in deep learning